Maintenance 4.0 combines Machine Learning–based Predictive Maintenance, Automation of Failure Reporting, Scheduling, Parts Management etc., and Robotics/Drone Assisted Repair.
The adoption of Maintenance 4.0 has broad implications for the industrial sector. We do not have to assume a crude ‘zero-sum game’ – that for every winner there will be a corresponding loser. At the same time, many stakeholders will struggle to adapt. While we do not expect disruption to occur overnight, some of the traditional players are vulnerable and may not survive the digitalization era.
OEMs: The Hardware as a Service Opportunity
“May you live in interesting times” captures the outlook for OEMs.
OEMs that are able to bundle industrial machinery with real-time Machine Learning–based Predictive Maintenance can migrate to a new business model – Hardware as a Service (HaaS).
A confluence of factors makes HaaS feasible. First, sensors that generate Big Data are already embedded in industrial machinery. Second, in the last few years, there has been a precipitous fall in the cost to capture, transport, store and analyze Big Data. By applying Machine Learning to sensor data, evolving equipment failure can be identified, and maintenance activities can be triggered prior to the actual breakdown.
With HaaS, OEMs can lease asset usage with maintenance. OEMs able to provide HaaS cost-effectively can shift to a higher–margin services model.
The challenge is that HaaS cannot be adopted with a short– to medium–term cost to the organization. Almost all aspects of the OEMs’ business will be impacted: Strategy, Process, Technology and People. There are risks associated with this change, and OEMs will need financial resources and operational agility to be successful.
The argument that over time HaaS is inevitable does not imply that all companies will be able to adopt it.
The challenge for many OEMs is a lack of resources and technical know-how to adapt to digitalization.
The Industrial Sector: The Overall Winners
In aggregate, Maintenance 4.0 is a big win for industrial producers, especially those in process industries.
To the extent that the various elements of Maintenance 4.0 – Machine Learning–based Predictive Maintenance, Automation, and Robotics/Drone Assisted Repair are adopted, industrial plants will benefit. This benefit is likely regardless of whether Operations and Maintenance (O&M) activities are executed by internal or external resources or whether industrial equipment is owned or leased.
The shift to Machine Learning-based Predictive Maintenance will result in lower overall maintenance costs and higher production yield rates. This is because four critical metrics (lost production revenue, repairs, waiting time and parts) all improve based on the shift from Reactive to Predictive Maintenance.
It should be noted that not all industrial plants are likely to benefit equally. Plants may lack the ability to be both agile and quick to adopt and scale the various elements of Maintenance 4.0. The extent to which these plants can readily access Big Data that is generated by sensors increases the likelihood of success.
In this regard, process industries that generate continuous streams of Big Data are at an advantage relative to batch process manufacturers that do not generate continuous data.
O&M Service Providers
Although a meaningful reduction in both expensive (and often unnecessary) Preventive Maintenance and Reactive Maintenance is expected to lower overall maintenance activities, we should not assume that O&M service providers will be hurt.
We expect confusion to be generated by a multitude of new technologies that make up Maintenance 4.0 solutions. O&M service providers that can assume a Trusted Advisory role and help plants navigate through the overwhelming technology options can maintain and grow their businesses.
Those that are considered transactional service providers and cannot add value beyond executing repair functions may lose to more adept competitors.
Maintenance 4.0 is attracting both new and existing players. Industry behemoths are being forced to invest in R&D, and will need to compete with Google, Microsoft and Amazon as well as venture capital-backed startups.
The move to Machine Learning is irreversible and not all legacy technologies and vendors will be able to adapt. Solutions that are labor intensive, such as statistical packages and SCADA monitoring, will be considerably enhanced by Machine Learning algorithms.
The data science field is in itself rapidly evolving as Automated Machine Learning (Auto-ML) is automating repetitive and time-consuming functions that were previously performed by data scientists.
With respect to technology vendors, the legacy system losers are easier to identify. However, with the significant investments in innovation, today’s winners may become obsolete.
Summary and Conclusion
Revolutions are unpredictable and chaotic. Apart from some solution providers of outdated technologies, we are unable to pick the Maintenance 4.0 winners and losers at this time.
In all four categories that we have covered – OEMs, Industrial Plants, Service Providers and Technology Vendors – new elites will emerge, some of the established players will lose out and others will transform themselves and thrive.
Industrial plants generate terabytes of process data. SKF Enlight AI is a SaaS Predictive Maintenance solution that uses Automated Machine Learning to identify emerging asset failure patterns within this data. It provides early warnings and sensor-level intelligence to help avert unplanned downtime and meet production goals. For more information on how SKF Enlight AI can improve performance and reliability, click here.